Continuous Object Region Detection in Collaborative Fog-Cloud IoT Networks

In a resource-constrained Internet of Things (IoT) networks, energy efficiency is a principle issue for monitoring the movement of continuous objects, such as wild fire and hazardous chemical material. These phenomena detection requires more reliable, in-situ techniques that can accurately adapt to nondeterministic and dynamic motion variation. Unfortunately, existing works that only focus on simple and well-defined shapes of phenomena are no longer sufficient. In this article, a continuous object monitoring scheme leveraging spatial grid index routing tree is proposed in IoT networks. To address the problem of energy consumption caused by a large number of exchanged messages, we put forward a novel detection mechanism based on probability density function and domination graph for identifying boundary points with the help of edge devices and cloud. Therefore, the proposed approach not only ensures high tracking accuracy, but also minimizes the number of exchanged messages involved in the detection and tracking process. Simulation results demonstrate that our detection approach can achieve higher tracking accuracy while significantly reduce the communication overhead compared to state-of-the-art methods.